Here is how I see the week's flow :
Table of content :
I've been working with a friend, Mauricio Cordoba, on his 1 dollar water filter project : FairCap.
FairCap was one of the 12 projects selected to participate in POC21.
Now the idea is to prototype a filter that adapts to plastic bottles and small pipes allowing at the same time the use of a new unexpensive porous tubing technology.
We'll mold and cast this part, the "multi-adapter" :
The porous tubes :
Here is a section of the whole system :
Again, we'll focus on the side multi-adapter
Here is our part to cast :
Before jumping on the design I read this guide, the guerilla guide to CNC machining.
Aside from the information that was given to us from the fabacademy course, there is some great explanations in it, like this one :
Also, doing the Solidworks tutorial on molding was good :
A great tool in Solidworks is the drafting analysis :
It allows us to see how the demoulding would go, and then help us doing slight angles on either part of the split between the two mold sides.
After tweaking the part, here is what shows the drafting tool :
Something that we also have to take into consideration when we design the mold is the angle that we need to make on the vertical faces for milling the deeper surfaces
Indeed, the bits need to have room to go down without colliding on other faces of the part.
In Solidworks, I do a quick sketch to find the angle based on measurements on the bit.
Going mentally through the process of demolding, I realize that the threads that are present on both side fo the part will make it impossible to demold by sliding with hard material for mold.
Not only this, but machining the thread with 3axis CNC seems to be impossible. We need to cast a mold part.
So two ideas :
I choose to go with the Silicon idea :
So here is an exploded view of the parts I need to make :
The silicone parts are mold parts that will be removed by squishing them and pulling them out.
I'm drawing positive parts that are simpler versions of the final part.
They will be used as positive molds for casting the silicon negative mold :
Then I will remove them and I'll be left with just the mold parts I need to cast the final part.
I just have to put the two sides together.
I give a 0.1 mm play for the right faces to be tight, not all of them :
I also make channels and vent for injecting the liquid plastic.
I guess it's better to fill from the bottom to the top to prevent trapped air.
Same thing for the other side :
Here is a video of how the mold works :
I start by 3Dprinting my positive parts (the black ones on the picture shown earlier).
One for the "green" side :
And for the red side :
The prints :
Then I generated the toolpath for machining the hard wax mold part.
For this, I need to export my geometry. A convenient thing to do is to create a coordinate system that outputs our part in the right orientation and position according to the software we use to generate the toolpath.
I use Rhino and RhinoCam to generate the toolpath.
Here are the machining jobs :
And the toolpaths that were generated with their simulation :
Just to check I also played with fabmodules :
It looks alright, I like how it's very simple to use.
I first need to cut a piece of wax that fits the mold part that I will mill :
Then I position it in our Monofab milling machine :
Here is a video of the first machining job, roughing :
The result :
The second operation is horizontal finishing. Here is the result :
Then we have the two parallel finishing operations.
On one of the mold parts, I also generated a toolpath and milled a channel with a 1/32" bit for injecting the plastic.
Here are the two parts at the end of the process :
The mold closes perfectly :
I pour the silicon in one of the mold :
I obtain the internal silicone negative :
As wished, I'm able to take out the silicone negative off of its 3D printed positive counterpart :
This Silicone part will then be the internal negative for our multiadapter.
I do the same for the other side :
Now, I just need to cast liquid plastic with these silicone internal negatives at the center of the two parts mold.
I didn't have the time to do this but you can check my final project page for a similar casting.
Here are the sources files of the projects I talked about on this page :
If you feel like contributing, please fork from Github :
...
***